# Forecasting Pedestrian Trajectory with Machine-Annotated Training Data

## TL;DR

This paper introduces a deep learning model for pedestrian trajectory forecasting using a large, machine-annotated dataset from a vehicle-mounted camera, improving prediction accuracy for autonomous driving applications.

## Contribution

We propose a scalable machine annotation scheme and a Dynamic Trajectory Predictor model trained on both human and machine-annotated data for improved pedestrian trajectory forecasting.

## Key findings

- The DTP model outperforms linear models in predicting pedestrian motion.
- Machine annotation enables training on large datasets without extensive human labeling.
- Experimental results validate the effectiveness of the proposed approach.

## Abstract

Reliable anticipation of pedestrian trajectory is imperative for the operation of autonomous vehicles and can significantly enhance the functionality of advanced driver assistance systems. While significant progress has been made in the field of pedestrian detection, forecasting pedestrian trajectories remains a challenging problem due to the unpredictable nature of pedestrians and the huge space of potentially useful features. In this work, we present a deep learning approach for pedestrian trajectory forecasting using a single vehicle-mounted camera. Deep learning models that have revolutionized other areas in computer vision have seen limited application to trajectory forecasting, in part due to the lack of richly annotated training data. We address the lack of training data by introducing a scalable machine annotation scheme that enables our model to be trained using a large dataset without human annotation. In addition, we propose Dynamic Trajectory Predictor (DTP), a model for forecasting pedestrian trajectory up to one second into the future. DTP is trained using both human and machine-annotated data, and anticipates dynamic motion that is not captured by linear models. Experimental evaluation confirms the benefits of the proposed model.